@article{article_1087926, title={Evaluation and comparison of metaheuristic methods for Markowitz’s mean-variance portfolio optimization model}, journal={İstatistikçiler Dergisi:İstatistik ve Aktüerya}, volume={15}, pages={19–33}, year={2022}, author={Yapıcı Pehlivan, Nimet and Yıldız, Berat}, keywords={Yapay arı kolonisi, Diferansiyel evrim, Metasezgiseller, Parçacık Sürü Optimizasyonu, Portföy optimizasyonu, Karesel programlama}, abstract={<p> Portfolio selection is the process of selecting a combination of assets among portfolios containing multiple assets to achieve a satisfactory return on investment. Mean-variance model proposed by Markowitz (1952) has been extensively used for portfolio selection problem. It is a quadratic programming model based on the minimum risk and maximum return by choosing assets in the portfolio. Generally, classical optimization algorithms have been used for solving the quadratic programming problem. Recently, metaheuristic optimization algorithms have been used in addition to classical optimization techniques for solving portfolio selection problems. Metaheuristic methods are designed to solve complex optimization problems that cannot be solved in a reasonable time with the definitive solution methods. Various metaheuristic methods have been developed for different areas. In this study, BIST30 index data set obtained from daily closing prices of 30 stocks between December 2016 - December 2017 was used. Markowitz’s mean-variance model is considered to constitute an optimal portfolio. , Particle Swarm Optimization, Differential Evolution, and Artificial Bee Colony which are mostly used metaheuristic methods, are applied to determine an optimal portfolio. Performances of these methods are compared by considering risk values, i.e. portfolio variances.   <br /> </p>}, number={1}, publisher={Aktüerya Derneği}